A Sparse Multiclass Motor Imagery EEG Classification Using 1D-ConvResNet
Multiclass motor imagery classification is essential for brain–computer interface systems such as prosthetic arms. The compressive sensing of EEG helps classify brain signals in real-time, which is necessary for a BCI system. However, compressive sensing is limited, despite its flexibility and data...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-03-01
|
| Series: | Signals |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2624-6120/4/1/13 |
| _version_ | 1797608974682423296 |
|---|---|
| author | Harshini Gangapuram Vidya Manian |
| author_facet | Harshini Gangapuram Vidya Manian |
| author_sort | Harshini Gangapuram |
| collection | DOAJ |
| description | Multiclass motor imagery classification is essential for brain–computer interface systems such as prosthetic arms. The compressive sensing of EEG helps classify brain signals in real-time, which is necessary for a BCI system. However, compressive sensing is limited, despite its flexibility and data efficiency, because of its sparsity and high computational cost in reconstructing signals. Although the constraint of sparsity in compressive sensing has been addressed through neural networks, its signal reconstruction remains slow, and the computational cost increases to classify the signals further. Therefore, we propose a 1D-Convolutional Residual Network that classifies EEG features in the compressed (sparse) domain without reconstructing the signal. First, we extract only wavelet features (energy and entropy) from raw EEG epochs to construct a dictionary. Next, we classify the given test EEG data based on the sparse representation of the dictionary. The proposed method is computationally inexpensive, fast, and has high classification accuracy as it uses a single feature to classify without preprocessing. The proposed method is trained, validated, and tested using multiclass motor imagery data of 109 subjects from the PhysioNet database. The results demonstrate that the proposed method outperforms state-of-the-art classifiers with 96.6% accuracy. |
| first_indexed | 2024-03-11T05:55:10Z |
| format | Article |
| id | doaj.art-4087853f32a349b1954ad5322b3eeff3 |
| institution | Directory Open Access Journal |
| issn | 2624-6120 |
| language | English |
| last_indexed | 2024-03-11T05:55:10Z |
| publishDate | 2023-03-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Signals |
| spelling | doaj.art-4087853f32a349b1954ad5322b3eeff32023-11-17T13:51:05ZengMDPI AGSignals2624-61202023-03-014123525010.3390/signals4010013A Sparse Multiclass Motor Imagery EEG Classification Using 1D-ConvResNetHarshini Gangapuram0Vidya Manian1Department of Bioengineering, University of Puerto Rico, Mayaguez, PR 00681-9000, USADepartment of Bioengineering, University of Puerto Rico, Mayaguez, PR 00681-9000, USAMulticlass motor imagery classification is essential for brain–computer interface systems such as prosthetic arms. The compressive sensing of EEG helps classify brain signals in real-time, which is necessary for a BCI system. However, compressive sensing is limited, despite its flexibility and data efficiency, because of its sparsity and high computational cost in reconstructing signals. Although the constraint of sparsity in compressive sensing has been addressed through neural networks, its signal reconstruction remains slow, and the computational cost increases to classify the signals further. Therefore, we propose a 1D-Convolutional Residual Network that classifies EEG features in the compressed (sparse) domain without reconstructing the signal. First, we extract only wavelet features (energy and entropy) from raw EEG epochs to construct a dictionary. Next, we classify the given test EEG data based on the sparse representation of the dictionary. The proposed method is computationally inexpensive, fast, and has high classification accuracy as it uses a single feature to classify without preprocessing. The proposed method is trained, validated, and tested using multiclass motor imagery data of 109 subjects from the PhysioNet database. The results demonstrate that the proposed method outperforms state-of-the-art classifiers with 96.6% accuracy.https://www.mdpi.com/2624-6120/4/1/13compressive sensingsparse representationmotor imagery EEG classification1D-CNNresidual networks |
| spellingShingle | Harshini Gangapuram Vidya Manian A Sparse Multiclass Motor Imagery EEG Classification Using 1D-ConvResNet Signals compressive sensing sparse representation motor imagery EEG classification 1D-CNN residual networks |
| title | A Sparse Multiclass Motor Imagery EEG Classification Using 1D-ConvResNet |
| title_full | A Sparse Multiclass Motor Imagery EEG Classification Using 1D-ConvResNet |
| title_fullStr | A Sparse Multiclass Motor Imagery EEG Classification Using 1D-ConvResNet |
| title_full_unstemmed | A Sparse Multiclass Motor Imagery EEG Classification Using 1D-ConvResNet |
| title_short | A Sparse Multiclass Motor Imagery EEG Classification Using 1D-ConvResNet |
| title_sort | sparse multiclass motor imagery eeg classification using 1d convresnet |
| topic | compressive sensing sparse representation motor imagery EEG classification 1D-CNN residual networks |
| url | https://www.mdpi.com/2624-6120/4/1/13 |
| work_keys_str_mv | AT harshinigangapuram asparsemulticlassmotorimageryeegclassificationusing1dconvresnet AT vidyamanian asparsemulticlassmotorimageryeegclassificationusing1dconvresnet AT harshinigangapuram sparsemulticlassmotorimageryeegclassificationusing1dconvresnet AT vidyamanian sparsemulticlassmotorimageryeegclassificationusing1dconvresnet |